spreadsheets
Pass
Audited by Gen Agent Trust Hub on Sep 22, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external spreadsheets via the 'read-range' and 'get' actions. Ingestion points: Spreadsheet contents are retrieved from the Google Sheets API and mapped to the agent's context. Boundary markers: The configuration does not define explicit delimiters or instructions to prevent the agent from interpreting spreadsheet data as commands. Capability inventory: The skill possesses capabilities to modify spreadsheet data ('append-values', 'update-values') and modify file permissions ('share'). Sanitization: There is no evidence of content filtering or sanitization performed on the data retrieved from the API.
- [EXTERNAL_DOWNLOADS]: The skill interacts with official Google domains (sheets.googleapis.com and www.googleapis.com) for its core functionality. These references to trusted services are considered safe and represent the intended behavior of the skill.
Audit Metadata